The n e w e ng l a n d j o u r na l of m e dic i n e

Spe ci a l R e p or t

Rapid Scaling Up of Covid-19 Diagnostic Testing

in the United States — The NIH RADx Initiative
Bruce J. Tromberg, Ph.D., Tara A. Schwetz, Ph.D., Eliseo J. Pérez‑Stable, M.D.,
Richard J. Hodes, M.D., Richard P. Woychik, Ph.D., Rick A. Bright, Ph.D.,
Rachael L. Fleurence, Ph.D., and Francis S. Collins, M.D., Ph.D.

The first reports of an unusual cluster of pneu-
monia cases in the city of Wuhan, China,
emerged in December 2019, heralding a global
pandemic. As of July 13, 2020, more than 3.3
million U.S. residents have received a diagnosis
of coronavirus disease 2019 (Covid-19), and more
than 135,000 have died.1 Of great concern are
the data showing the disproportionate effect of
Covid-19 on ethnic and racial minorities.2,3 Since
January 2020, the National Institutes of Health
(NIH) has been involved in multiple wide-rang-
ing collaborative efforts spanning the develop-
ment of vaccines and diagnostic strategies, the
identification and evaluation of safe and effec-
tive treatments, the understanding of the natural
history of the disease, and the study of racial
and ethnic disparities.4 In this article, we de-
scribe the additional role of the NIH in the effort
to increase the range and availability of diagnos-
tic tests for the causative virus, SARS-CoV-2
(severe acute respiratory syndrome coronavirus 2).
We begin with a review of current and pro-
jected testing capacity needs and review differ-
ent types of diagnostic tests. We then describe
the Rapid Acceleration of Diagnostics (RADx)
program, its goals, and its focus on underserved
populations. As will become clear, this program
represents a dramatic extension of the usual
NIH mode of supporting research. RADx was
established in just a few days; it covers the entire
life cycle of the target technologies; it is tightly
focused on timelines and outcomes; it receives
applications primarily from small companies; it
is partnering with other agencies such as the
Office of the Assistant Secretary for Health, the
Biomedical Advanced Research and Develop-
ment Authority (BARDA), and the Department of
Defense; and it is expressly focused on health
disparities. We describe here the four components
of RADx and their goals, and we end with a re-
view of the challenges ahead.
On April 24, 2020, Congress appropriated
$1.5 billion, from the $25 billion provided in the
Paycheck Protection Program and Health Care
Enhancement Act for SARS-CoV-2 testing, to the
NIH. Within 5 days after the legislation was
signed into law, the NIH launched RADx to sup-
port the development, production scale-up, and
deployment of accurate, rapid tests across the
country. From a timing perspective, the RADx
initiative was conceived by Congress to provide
near-term solutions to increase the number of
tests available by the fall of 2020, as schools and
universities evaluate the safety of in-person
classes and as the annual influenza season be-
gins. In the slightly longer term, RADx also
aims to support the development and production
of innovative diagnostic technologies as well as
strategies for making testing available to di-
verse, vulnerable, and underserved populations
through 2021. One of the goals of the RADx
initiative is to expand capacity so that by Decem-
ber 2020, approximately 2% of the U.S. popula-
tion (6 million persons) can be tested per day,
with more tests ready for rapid deployment in
proportion to national demand.
Current Te s ting C apacit y
and Pr ojec tions
In the week leading up to July 13, 2020, daily
diagnostic testing capacity in the United States
was fluctuating between 520,000 and 823,000
tests.5 Models that provide robust estimates of
the number of tests needed per day vary widely.
Some experts estimated that 900,000 tests per
day would be needed in May.6 Others forecasted
the need for 5 million tests per day by June, in-

n engl j med 383;11  nejm.org  September 10, 2020 1071

The New England Journal of Medicine
Downloaded from nejm.org on September 11, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l
creasing to 20 million tests per day by July.7 Al-
though national totals are helpful benchmarks,
the models must account for a range of vari-
ables, including different levels of regional prev-
alence and community spread, the needs of high-
risk communities (e.g., nursing homes, shelters,
prisons, and factories), and the frequencies and
types of testing to be conducted. For example,
the type and characteristics of a test that are
required will differ when testing is conducted to
determine personal infection status as compared
with evaluating population-level surveillance. Test
performance (the limit of detection, sensitivity,
specificity, and the positive predictive value),
turnaround time, cost, accessibility, and accep-
tance are critical factors in a successful testing
strategy that can meet the needs of individual
persons and communities across the country.
T ype s of Diagnos tic Te s t s
Current diagnosis of acute SARS-CoV-2 infection
relies on tests that detect either viral RNA or viral
antigens.8 Most existing methods for Covid-19
testing use reverse-transcriptase–polymerase-
chain-reaction (RT-PCR) tests that detect nucleic
acid sequences specific to SARS-CoV-2. These
tests are highly sensitive and specific when con-
ducted in centralized laboratories with standard-
ized protocols, but they require a large amount
of laboratory space, complex equipment, regula-
tory approvals for the laboratory operations, and
skilled laboratory leadership and technicians.
Results generally take time to become available,
with windows ranging from hours to days, and
the need for transport of specimens to a central
laboratory leads to further delays. For this rea-
son, low-complexity molecular diagnostic point-
of-care tests with rapid turnaround have sub-
stantial practical advantages.9 A number of
point-of-care tests have now received Emergency
Use Authorization (EUA) by the Food and Drug
Administration (FDA).10
Antigen tests work by detecting the presence
of viral proteins and can provide rapid results,
similar to the way pregnancy tests operate.9 Anti-
gen tests can offer fast and scalable point-of-care
performance, but they have lower sensitivity than
most nucleic acid–based assays.11 This limita-
tion can be a concern in high-risk settings such
as nursing homes, where missing the detection
of an infected person can lead to serious conse-
1072 n engl j med 383;11  nejm.org  September 10, 2020
The New England Journal of Medicine
Downloaded from nejm.org on September 11, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
of m e dic i n e
quences. Although a number of manufacturers
are known to be developing antigen-based tests,
to date only two have received an FDA EUA.12,13
Serologic tests that detect antibody response
are also being developed by numerous compa-
nies, but these tests are not suitable for the di-
agnosis of acute infection, since human antibod-
ies are not formed until 2 to 3 weeks after viral
infection. Thus, their primary use is to docu-
ment previous exposure to the virus.14 The devel-
opment and scaling of serologic tests are not
under the auspices of the NIH RADx program.
Comp onent s of the R AD x Pr o gr am
The RADx program has four components. RADx-
tech aims to identify, accelerate the development
of, scale up, and deploy innovative point-of-care
technologies as early as the fall of 2020. RADx–
Advanced Technology Platforms (RADx-ATP) will
support the scale-up of somewhat more ad-
vanced technologies that can achieve immediate,
substantial increases in capacity. RADx-rad (short-
hand for radical) will focus on truly nontradi-
tional approaches for testing that have a slightly
longer horizon. RADx–Underserved Populations
(RADx-UP) will establish community-engaged
implementation projects to improve access to test-
ing in underserved and vulnerable populations.
Le ver aging Our Scientific
Cre ativit y — R AD x -tech
The RADx-tech program offers extensive exper-
tise and support to bring the diagnostic tech-
nologies that require additional clinical, regula-
tory, and commercialization assistance from
development to deployment. The RADx-tech pro-
gram uses a rigorous, rapid-review process that
provides independent evaluation of the technol-
ogy and the potential to scale. The program le-
verages the long-standing Point-of-Care Technol-
ogy Research Network (POCTRN), which is run
by the National Institute of Biomedical Imaging
and Bioengineering.15 In a process based on an
“innovation funnel,” applications move rapidly
through multiple review gates that involve in-
creasing selection pressure (Fig. 1).
On entering the innovation funnel, each proj-
ect is evaluated by a team with wide-ranging ex-
pertise. Projects that are deemed to be promising
enter into to a weeklong intensive review pro-
 Special Report

National Call
for Innovative End of Summer
Technologies Phase 0 Phase 1 Phase 2 or Fall 2020
Rolling submissions “Shark tank”–like Validation and Clinical tests,
and selections begin rapid-selection risk review regulatory approval,
April 29, 2020 process and scaling up

Fast Track for Advanced Technologies: the RADx-ATP Program

Deploy Millions
of Tests per Week

Figure 1. RADx-tech Innovation Funnel to Evaluate Testing Technologies for Covid-19.

Applications to the Rapid Acceleration of Diagnostics (RADx)–tech program for the detection of severe acute respi‑
ratory syndrome coronavirus 2 (SARS‑CoV‑2), which causes coronavirus disease 2019 (Covid‑19), move rapidly through
an “innovation funnel” from phase 0 to phase 1 to phase 2, with increasing selection pressure at each gate. The
RADx–Advanced Technologies Platforms (RADx‑ATP) program identifies projects that are more advanced and that
can be moved immediately to phase 1 or even phase 2.

cess, which we refer to as a “shark tank” or

“deep dive.”16 In phase 0, multiple expert review-
ers provide a detailed assessment of the technol- Small Business
(<50 employees)
ogy and give critical feedback to the NIH and
the project teams. Approximately 15 to 20% of Academic Laboratory
completed RADx-tech applications enter phase 0.
For those projects that are successful (25 to 30% Start-up (<1 yr)
of phase 0 projects), detailed, milestone-driven
work packages are developed in order for proj- Midsize Business
ects to enter phase 1. Technologies are then (50–250 employees)
rigorously tested and validated in the indepen- Large Business
dent POCTRN validation core over a monthlong (>250 employees)
process to ensure that these new tests meet or
Other
exceed their predicted analytic performance. If a
project is judged to be successful at that point, Nonprofit Laboratory
rapid scale-up and clinical testing in phase 2 gets or Contract Research
Organization
under way, with substantial financial assistance
provided. 0 50 100 150 200 250
Applications that have been received to date No. of Proposals
by the RADx-tech program span nearly every
stage of technology development, with submis- Figure 2. Number of RADx-tech Proposals, According to Organization Type.
sions coming from small and midsize compa- The majority of applications to the RADx‑tech and the RADx‑ATP programs
nies, academic laboratories, early-stage start-up have come from small businesses, academic laboratories, and early‑stage
companies, and large commercial manufacturers start‑up companies.
(Fig. 2). It is not uncommon for applicants to

n engl j med 383;11 nejm.org September 10, 2020 1073

The New England Journal of Medicine
Downloaded from nejm.org on September 11, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

phase 1, and the entry of the first project into

Not specified,
Feces, 1.4% phase 2 is imminent. As technologies and com-
2.8% panies go through this process, the RADx pro-
gram is also seeking to identify testing plat-
forms that may be especially suitable for testing
Blood, small groups or isolated, underserved popula-
8.8%
Saliva, tions at point-of-care sites or in rural or remote
24.5% areas that do not have access to high-throughput
robotic testing systems. Technologies that reduce
Other,
10.1% the facility footprint, decrease overall testing
complexity, and provide rapid results will be es-
pecially helpful, and the program will work to
assist in the deployment of these systems in these
specialized environments. As part of this pro-
Sputum, cess, ease of use, including specimen-collection
11.9%
methods and clear and easy-to-understand in-
Nasal swab,
structions to facilitate widespread uptake, will
22.6% be integrated into the process.
Active collaborations with other government
Oral swab, agencies are critical during this process. The
17.9%
NIH is closely coordinating with the Office of
the Assistant Secretary for Health, BARDA, and
Department of Defense to ensure that each party
is aware of the discussions and existing relation-
Figure 3. Sample Types in RADx-tech and RADx-ATP Proposals.
ships taking place with private-sector compa-
Applications to RADx-tech and RADx-ATP include multiple alternative sam‑
pling strategies for viral testing, not just traditional nasal swabs. nies, to ensure no duplication of funding, to
streamline communications, and to leverage the
experience, knowledge, and technical capacities
still be preparing for their EUA submission to of different agencies within the government.
the FDA. Promising technologies that have been
reviewed by the program include innovations Achie ving Shor t-Term, R apid
that allow high-throughput, portable, and point- Sc ale - up — R AD x - ATP
of-care platforms, from CRISPR (clustered regu-
larly interspaced short palindromic repeats) tech- Not all emerging technologies need the shark
nologies for the detection of viral nucleic acids tank approach. As shown in Figure 1, a bypass
to lateral flow strips for both nucleic acid and pathway is also provided to catalyze the rapid
viral antigen testing.17 Many platforms integrate development of technologies that are already
advanced microfluidic components and state-of at a more advanced stage of development. The
the-art readout strategies that use compact op- RADx–Advanced Technologies Platforms (RADx-
tics and electronics. Overall performance, ease of ATP) program provides a rapid-response applica-
use, and digital reporting are facilitated by smart- tion process for companies with an existing
phone systems that are designed for nonexperts. point-of-care technology that has already been
In addition, alternatives to conventional naso- authorized by the FDA for the detection of SARS-
pharyngeal and anterior nasal swab sampling CoV-2 and that has the ability to scale produc-
are being explored,18 and a majority of proto- tion to between 20,000 and 100,000 tests per day
types involve the use of saliva, oral swabs, and by the fall of 2020. In addition, the company
other collection sites (Fig. 3). must have infrastructure and a deployment or
As of July 13, 2020, a total of 2587 expres- placement strategy that enables tests to be made
sions of interest have been received, and more rapidly available in point-of-care settings. Com-
than 600 full applications have been submitted panies are asked to develop a robust plan to
from 41 U.S. states and the District of Columbia. collect data continuously to support a submis-
Within 8 weeks after the launch, 27 projects had sion to the FDA for a Premarket Approval (PMA)
successfully made it through the shark tank to or a Premarket Notification (510[k]).

1074 n engl j med 383;11  nejm.org  September 10, 2020

The New England Journal of Medicine

Downloaded from nejm.org on September 11, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
 Special Report
The RADx-ATP program is also seeking to
expand high-throughput laboratories (also called
“mega-labs”) that are in a position to increase
testing capacity to 100,000 to 250,000 tests per
day. These laboratories have been certified ac-
cording to the Clinical Laboratory Improvement
Amendments regulations and are already in op-
eration with the necessary equipment and trained
staff to guarantee a test-turnaround time of 24
hours (from the time that the sample is obtained
to the availability of the result). Such high-
throughput laboratories generally have the equip-
ment to analyze large numbers of tests with
well-developed automated workflows to process
samples and obtain results rapidly. Newer tech-
nologies, such as next-generation sequencing,
can read out hundreds to thousands of genes or
gene regions simultaneously19 and should be able
to support large-scale, population-level testing
for surveillance purposes, possibly on the order
of millions per day.
Pooling strategies are also being considered
for population-level surveillance, because they
can greatly increase throughput when testing
resources are limited.20 If a pooled test result is
negative, then the patients who provided the
samples are considered to be negative for active
SARS-CoV-2 infection. If a pooled test result is
positive, then each sample from the pool of pa-
tients needs to be tested individually. The effi-
ciency of pooling has been shown to depend on
the prevalence of SARS-CoV-2, test sensitivity,
and patient-pool size and will need further
evaluation before widespread implementation.21
Molecular bar-coding is another promising ap-
proach. In the first step, the individual sample is
labeled with a DNA tag, in order for it to be
identifiable at the molecular level. Massive pool-
ing can then be done to increase the number of
tests that can be processed at once. Positive
samples can be identified by their unique tag,
but the pooling step can greatly decrease the use
of reagents, equipment, and labor.22 This approach
may even allow for a range of other viral patho-
gens to be detected during the same analysis.
Lo oking Fur ther Ahe ad
— R AD x - r ad
Not all technologies will be ready for near-term
production scale-up and deployment. A special
component of the program (RADx-rad) has been
established to evaluate the usability, access, ro-
n engl j med 383;11  nejm.org  September 10, 2020
The New England Journal of Medicine
Downloaded from nejm.org on September 11, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
bustness, or accuracy of a wide range of nontra-
ditional technologies or new settings (e.g., home-
based testing technologies) for the detection of
SARS-CoV-2 infections. Projects under the RADx-
rad component will be focused not just on the
development of new technologies or approaches
but also on the novel repurposing of existing
technologies. Furthermore, RADx-rad will sup-
port those innovative technologies that are on a
more extended timescale and that will take lon-
ger to develop than a 6-month time frame. Such
technologies include, for example, the use of
biologic or physiological biomarkers to detect an
infection or predict the severity of disease,23
including the likelihood of the multisystem in-
flammatory syndrome in children,24 or the use
of chemosensory changes as an early indicator of
viral positivity.25 Other examples include the use
of biosensors to detect the presence of the virus
in the breath26 or the analysis of wastewater to
conduct community-based surveillance.27
Fo cus on Under ser ved
P opul ations — R AD x - UP
It is clear that racial and ethnic minorities are
bearing a higher burden of disease and mortal-
ity from Covid-19.3 In particular, non-Hispanic
Blacks, Hispanics, and American Indians and
Alaska Natives are hospitalized and die at dis-
proportionately higher rates than other groups.28-31
This disproportionate burden in health outcomes
for underserved populations and racial and eth-
nic minorities shines a bright light on long-
standing health disparities in the United States
and is of profound concern.32
The goal of this part of the RADx program
(RADx-UP) is to understand factors that have led
to the disproportionate burden of the pandemic
on underserved populations and to support im-
proved access and uptake of SARS-CoV-2 testing.
The program aims to examine infection patterns
and efforts in order to increase access to and
effectiveness of testing methods by building an
infrastructure that can be leveraged for the on-
going Covid-19 public health efforts, especially
as the impending influenza season begins in the
fall. Through engagement with communities and
in close alignment with their leaders, a series of
interlinked, pragmatic implementation science
projects at multiple sites across the country are
being planned to investigate, in real time, the
effective approaches to testing within these
1075
 The n e w e ng l a n d j o u r na l
populations. In addition, in order to be respon-
sive to the communities that this initiative will
partner with and to understand more clearly the
multitude of factors influencing the ability and
willingness of a group to be tested for SARS-
CoV-2, the program is obtaining strong guid-
ance on social, ethical, and behavioral issues by
means of the establishment of a research pro-
gram focused solely on these issues.
Challenge s Ahe ad
The NIH has been engaged almost continuously
in battling epidemic diseases over the past sev-
eral decades. From human immunodeficiency vi-
rus (HIV) infection to influenza, SARS, Middle
East respiratory syndrome (MERS), Ebola, and
Zika, each pandemic brings its own idiosyn-
cratic issues that require unique solutions. We
anticipate that challenges to the RADx program
are likely to arise on several fronts. First, on the
technical side, many promising prototypes un-
der consideration are in the early proof-of-con-
cept stage. It is likely that many will fail along
the way. Much effort will be needed to conduct
clinical validation rapidly and to obtain regula-
tory authorizations and approvals. Access to
clinical samples for clinical validation has al-
ready emerged as an issue that is impeding
progress for companies.
Another challenge will be to identify digital
health platforms that provide connectivity among
test results, electronic health records, and public
health organizations. Many smaller diagnostic
companies do not have expertise in this area,
and it will be important to ensure their access to
this expertise in the planning stages. Data should
be able to be transmitted to the patient and
health care provider and to state and local public
health authorities with the use of national inter­
operability standards and common data elements
to collect key demographic, test result, and
clinical outcome information.33 This will lay the
foundation for the future point-of-care clinical
practice and research enterprise in the years to
come, in which deidentified, privacy-protected,
patient-level test data can be linked to clinical
outcomes to form curated, analyzable data sets.
Second, on the manufacturing side, scaling up
of production is a complex enterprise that in-
volves a wide range of factors, from ensuring
1076 n engl j med 383;11  nejm.org  September 10, 2020
The New England Journal of Medicine
Downloaded from nejm.org on September 11, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.
of m e dic i n e
sufficient supplies of swabs, raw materials, re-
agents, and equipment to establishing new pro-
duction lines and manufacturing facilities with-
in aggressive timelines. These are likely to be
logistic feats that will require rapid access to
capital, equipment, and skilled staff.
Third, on the distribution side, once in-
creased testing capacity is available, managing
the distribution and implementation of tests into
the appropriate venues and geographic localities
will be critical. We anticipate coordinating
closely with Operation Warp Speed34 and lever-
aging the logistic expertise of the Department of
Defense. Uptake of testing may also be an issue.
All these challenges are being addressed with
unprecedented levels of coordination and col-
laboration across academia, government, in-
dustry, and nonprofit foundations.35 We antici-
pate that these interactions will substantially
affect traditional barriers and facilitate the first
burst of increased testing capacity by the fall
of 2020.
Conclusions
Expanding the capacity, throughput, speed of
returning results, analytic performance, and re-
gional placement of diagnostic technologies is
urgently needed and, if successful, will contrib-
ute importantly to the current national efforts to
curb the Covid-19 pandemic and help to reduce
inequities for underserved populations. As we
embark on this initiative, the challenges ahead
are considerable, and the timetable is truly
daunting. Aiming to achieve this rapid evalua-
tion, validation, and scale-up has rarely, if ever,
been attempted at this pace. However, the NIH
is in a position to serve as a “venture invest-
ment” organization and is currently striving to
operate in that entrepreneurial spirit. The suc-
cess of the RADx program will depend on truly
innovative ideas coming forward from the minds
and laboratories of technology developers, a ro-
bust and rapidly responsive expert evaluation
system, extensive collaborations in validation
and scale-up with experts from all sectors, and
strong community partnerships to support test-
ing availability and uptake. All these partners
are profoundly energized by a sense of urgency,
opportunity, and responsibility to provide test-
ing at scale in the face of this global pandemic.
 Special Report

Disclosure forms provided by the authors are available with 17. Broughton JP, Deng X, Yu G, et al. CRISPR-Cas12-based de-
the full text of this article at NEJM.org. tection of SARS-CoV-2. Nat Biotechnol 2020;​38:​870-4.
18. Tu Y-P, Jennings R, Hart B, et al. Swabs collected by patients
From the National Institute of Biomedical Imaging and Bioen‑ or health care workers for SARS-CoV-2 testing. N Engl J Med.
gineering (B.J.T.), the National Institute of Nursing Research DOI:​ 10.1056/NEJMc2016321.
(T.A.S.), the Office of the Director (T.A.S., R.A.B., R.L.F., 19. Gwinn M, MacCannell D, Armstrong GL. Next-generation
F.S.C.), the National Institute on Minority Health and Health sequencing of infectious pathogens. JAMA 2019;​321:​893-4.
Disparities (E.J.P.-S.), the National Institute on Aging (R.J.H.), 20. Hogan CA, Sahoo MK, Pinsky BA. Sample pooling as a strat-
and the National Institute of Environmental Health Sciences, egy to detect community transmission of SARS-CoV-2. JAMA
National Toxicology Program (R.P.W.), National Institutes of 2020;​323:​1967-9.
Health, Bethesda, MD. Address reprint requests to Dr. Collins 21. Cherif A, Grobe N, Wang X, Kotanko P. Simulation of pool
at the Office of the Director, National Institutes of Health, testing to identify patients with coronavirus disease 2019 under
1 Center Dr., Bldg. 1, Rm. 126, Bethesda, MD 20892, or at conditions of limited test availability. JAMA Netw Open 2020;​
­collinsf@​­od​.­nih​.­gov. 3(6):​e2013075.
This article was published on July 22, 2020, at NEJM.org. 22. Schmid-Burgk JL, Li D, Feldman D, et al. LAMP-Seq:​popula-
tion-scale COVID-19 diagnostics using a compressed barcode
1. Center for Systems Science and Engineering at Johns Hop- space. June 8, 2020 (http://biorxiv​.org/​content/​early/​2020/​04/​08/​
kins University. COVID-19 dashboard. 2020 (https://coronavirus​ 2020​.04​.06​.025635​.abstract). preprint.
.jhu​.edu/​map​.html). 23. Russell SM, Alba-Patiño A, Barón E, Borges M, Gonzalez-
2. Webb Hooper M, Nápoles AM, Pérez-Stable EJ. COVID-19 Freire M, de la Rica R. Biosensors for managing the COVID-19
and racial/ethnic disparities. JAMA 2020 May 11 (Epub ahead of cytokine storm: challenges ahead. ACS Sens 2020;​5:​1506-13.
print). 24. Centers for Disease Control and Prevention. Emergency pre-
3. Stokes EK, Zambrano LD, Anderson KN, et al. Coronavirus paredness and response:​multisystem inflammatory syndrome
disease 2019 case surveillance — United States, January 22–May in children (MIS-C) associated with coronavirus disease 2019
30, 2020. MMWR Morb Mortal Wkly Rep 2020;​69:​759-65. (COVID-19). May 14, 2020 (https://emergency​.cdc​.gov/​han/​2020/​
4. Collins FS, Stoffels P. Accelerating COVID-19 Therapeutic han00432​.asp).
Interventions and Vaccines (ACTIV): an unprecedented partner- 25. Yan CH, Faraji F, Prajapati DP, Boone CE, DeConde AS. Asso-
ship for unprecedented times. JAMA 2020 May 18 (Epub ahead ciation of chemosensory dysfunction and COVID-19 in patients
of print). presenting with inf luenza-like symptoms. Int Forum Allergy
5. The COVID Tracking Project. US historical data 2020 Rhinol 2020;​10:​806-13.
(https://covidtracking​.com/​data/​us​-­daily). 26. Gould O, Ratcliffe N, Krol E, de Lacy Costello BPJ. Breath
6. Jha AK, Jacobson B, Friedhoff S, Tsai T. HGHI and NPR pub- analysis for detection of viral infection, the current position of
lish new state testing targets. Harvard Global Health Institute. the field. J Breath Res 2020 June 12 (Epub ahead of print).
May 7, 2020 (https://globalepidemics​.org/​2020/​05/​07/​hghi​-­projected​ 27. Peccia J, Zulli A, Brackney DE, et al. SARS-CoV-2 RNA con-
-­tests​-­needed​-­may15/​). centrations in primary municipal sewage sludge as a leading
7. Roadmap to pandemic resilience. Cambridge, MA:​Edmond J. indicator of COVID-19 outbreak dynamics. June 12, 2020 (http://
Safra Center for Ethics at Harvard University, April 20, 2020 (https:// medrxiv​.org/​content/​e arly/​2020/​05/​22/​2020​.05​.19​.20105999​
ethics​.harvard​.edu/​f iles/​center​-­for​-­ethics/​f iles/​roadmaptopandemic .abstract). preprint.
resilience_updated_4​.20​.20_0​.pdf). 28. Price-Haywood EG, Burton J, Fort D, Seoane L. Hospitaliza-
8. Cheng MP, Papenburg J, Desjardins M, et al. Diagnostic test- tion and mortality among black patients and white patients with
ing for severe acute respiratory syndrome-related coronavirus 2: Covid-19. N Engl J Med 2020;​382:​2534-43.
a narrative review. Ann Intern Med 2020;​172:​726-34. 29. Martinez DA, Hinson JS, Klein EY, et al. SARS-CoV-2 positiv-
9. Sheridan C. Fast, portable tests come online to curb corona- ity rate for Latinos in the Baltimore–Washington, DC Region.
virus pandemic. Nat Biotechnol 2020;​38:​515-8. JAMA 2020 June 18 (Epub ahead of print).
10. Food and Drug Administration. Emergency use authoriza- 30. Artiga S, Orgera K. COVID-19 presents significant risks for
tions for medical devices:​EUAs for coronavirus disease 2019 American Indian and Alaska Native People. San Francisco:​Kaiser
(COVID-19). 2020 (https://www​.fda​.gov/​medical​-­devices/​emergency​ Family Foundation, May 14, 2020 (https://www​.kff​.org/​coronavirus​
-­situations​-­medical​-­devices/​emergency​-­use​-­authorizations). -­covid​-­19/​issue​-­brief/​covid​-­19​-­presents​-­significant​-­r isks​-­for​
11. Prendergast C, Papenburg J. Rapid antigen-based testing for -­a merican​-­indian​-­and​-­a laska​-­native​-­people/​).
respiratory syncytial virus: moving diagnostics from bench to 31. Centers for Disease Control and Prevention. Coronavirus
bedside? Future Microbiol 2013;​8:​435-44. disease 2019:​racial & ethnic minority groups. 2020 (https://
12. Food and Drug Administration. Quidel emergency use au- www​.cdc​.gov/​coronavirus/​2019​-­ncov/​need​-­extra​-­precautions/​racial​
thorization (EUA). 2020 (https://www​.fda​.gov/​media/​137886/​ -­ethnic​-­minorities​.html).
download). 32. Chowkwanyun M, Reed AL Jr. Racial health disparities
13. Food and Drug Administration. Emergency use authoriza- and Covid-19 — caution and context. N Engl J Med 2020;​383:​
tion to Becton, Dickinson and Company for BD Veritor System 201-3.
for rapid detection of SARS-CoV-2. July 2, 2020 (https://www​.fda​ 33. Sittig DF, Singh H. COVID-19 and the need for a national
.gov/​media/​139752/​download). health information technology infrastructure. JAMA 2020 May
14. Sethuraman N, Jeremiah SS, Ryo A. Interpreting diagnostic 18 (Epub ahead of print).
tests for SARS-CoV-2. JAMA 2020;​323:​2249-51. 34. Department of Health and Human Services. Fact sheet:​ex-
15. Ford Carleton P, Schachter S, Parrish JA, et al. National In- plaining Operation Warp Speed. June 16, 2020 (https://www​.hhs​
stitute of Biomedical Imaging and Bioengineering Point-of-Care .gov/​about/​news/​2020/​06/​16/​fact​-­sheet​-­explaining​-­operation​-­warp​
Technology Research Network: advancing precision medicine. -­speed​.html).
IEEE J Transl Eng Health Med 2016;​4:​2800614. 35. Zeidel ML, Kirk C, Linville-Engler B. Opening up new supply
16. Alexander L, Blunt R. We need more covid-19 tests. We pro- chains. N Engl J Med 2020;​382(21):​e73.
pose a ‘shark tank’ to get us there. Washington Post. April 20,
2020 (https://www​.washingtonpost​.com/​opinions/​2020/​04/​20/​how​ DOI: 10.1056/NEJMsr2022263
-­speed​-­up​-­testing​-­shark​-­t ank​-­government/​). Copyright © 2020 Massachusetts Medical Society.

n engl j med 383;11  nejm.org  September 10, 2020 1077

The New England Journal of Medicine
Downloaded from nejm.org on September 11, 2020. For personal use only. No other uses without permission.
Copyright © 2020 Massachusetts Medical Society. All rights reserved.